The pulmonary contusion score: Development of a simple scoring system for blunt lung injury

Abstract

Background

Pulmonary contusions (PC) are common after blunt chest trauma and can be identified with computed tomography (CT). Complex scoring systems for grading PC exist, however recent scoring systems rely on computer-generated algorithms that are not readily available at all hospitals. We developed a scoring system for grading PC to predict the need for prolonged mechanical ventilation and initial hospital admission location.

Methods

A retrospective review was performed of adult blunt trauma patients with PC identified on initial chest CT during 2020. Data elements related to demographics, injury characteristics, disposition and healthcare utilization were extracted. The primary outcome was the need for mechanical ventilation for greater than 48 h. A novel scoring system, the Pulmonary Contusion Score (PCS) was developed. The maximum score was 10, with each lobe contributing up to 2 points. A score of 0 was given for no contusion present in the lobe, 1 for less than 50 % contusion, and 2 for greater than 50 % contusion. A PCS of 4 was hypothesized to correlate with need for mechanical ventilation for over 48 h. A confusion matrix of the scoring algorithm was created, and inter-rater concordance was calculated from a randomly selected 125 patients.

Results

A total of 217 patients were identified. 118 patients (54 %) were admitted to the ICU, but only 23 patients (19 %) were intubated, and only 17 patients (8 %) required mechanical ventilation > 48 h. Sensitivity of the scoring system was 20 %, while specificity was 93 %. Negative predictive value was 93 %. Inter-rater agreement was 77 %.

Conclusion

The PCS is a scoring system with high specificity and negative predictive value that can be used to evaluate the need for mechanical ventilation after sustaining blunt PC and can help properly allocate hospital resources.

Level of evidence

IV - diagnostic criteria